Method for localization of nodes by using partial order of the nodes
First Claim
1. A system operable to localize nodes (X, . . . , Y) communicating in a wireless network, wherein each node (X, . . . , Y) comprises a partial ordering means operable to initially classify its node (X, . . . , Y) as an anchor node (A) if a measure of uncertainty of the state estimate, such as a state estimate error covariance norm, is below a first threshold value, or else classify its node (X, . . . , Y) as a non-anchor node, giving rise to a partial order of said nodes (X, . . . , Y), wherein each node (X, . . . , Y) also comprises a control means connected to said partial ordering means, a to said control means connected recursive filter means, wherein every time a node (X) receives a message from another node (Y) said control means is operable to check that said node (Y) precedes said node (X) in said partial order, wherein a node (Y) preceding a node (X) meaning that said node (Y) being located in an ordered sequence comprising both node (X) and node (Y) and being located closer than node (X) to an initial anchor node in said sequence, whereby said recursive filter means is operable to be applied on said current state and measurement, giving an updated state, whereby said partial ordering means, if node (Y) precedes node (X), is operable to update said status of said node (X) in said partial order, wherein each control means is operable to repeat the above, giving the position and orientation of each node (X, . . . , Y), and in that for said system at least one estimation is chosen from the group consisting of the two estimations,position estimation, where the measurement equation system comprises vectors y and x related as Gy=Gx, where G is a matrix, where y is a vector comprising a position coordinate estimate for a node (Y) received from said node (Y), and where x is a state vector comprising a position coordinate for node (X), andorientation estimation, where the measurement equation system comprises vectors z and x related as z=Hx, where H is a matrix, where z is a measurement vector comprising an angle of arrival for node (X), and where x is a state vector comprising an orientation coordinate and an angle of arrival for node (X).
2 Assignments
0 Petitions
Accused Products
Abstract
The present invention relates to a new method for localization, i.e. finding positions and orientations of nodes communicating in wireless networks. The method is based on using directional information (angle-of-arrival, AOA), and optionally distance information, combined with estimation by recursive filters such as e.g., Kaiman filters, recursive least squares filters, Bayesian filters, or particle filters. The method expresses the localization problem as set of linear equations and ensures stability and convergence by imposing a partial order on the nodes.
-
Citations
19 Claims
-
1. A system operable to localize nodes (X, . . . , Y) communicating in a wireless network, wherein each node (X, . . . , Y) comprises a partial ordering means operable to initially classify its node (X, . . . , Y) as an anchor node (A) if a measure of uncertainty of the state estimate, such as a state estimate error covariance norm, is below a first threshold value, or else classify its node (X, . . . , Y) as a non-anchor node, giving rise to a partial order of said nodes (X, . . . , Y), wherein each node (X, . . . , Y) also comprises a control means connected to said partial ordering means, a to said control means connected recursive filter means, wherein every time a node (X) receives a message from another node (Y) said control means is operable to check that said node (Y) precedes said node (X) in said partial order, wherein a node (Y) preceding a node (X) meaning that said node (Y) being located in an ordered sequence comprising both node (X) and node (Y) and being located closer than node (X) to an initial anchor node in said sequence, whereby said recursive filter means is operable to be applied on said current state and measurement, giving an updated state, whereby said partial ordering means, if node (Y) precedes node (X), is operable to update said status of said node (X) in said partial order, wherein each control means is operable to repeat the above, giving the position and orientation of each node (X, . . . , Y), and in that for said system at least one estimation is chosen from the group consisting of the two estimations,
position estimation, where the measurement equation system comprises vectors y and x related as Gy=Gx, where G is a matrix, where y is a vector comprising a position coordinate estimate for a node (Y) received from said node (Y), and where x is a state vector comprising a position coordinate for node (X), and orientation estimation, where the measurement equation system comprises vectors z and x related as z=Hx, where H is a matrix, where z is a measurement vector comprising an angle of arrival for node (X), and where x is a state vector comprising an orientation coordinate and an angle of arrival for node (X).
-
10. A method for localization of nodes (X, . . . , Y) communicating in a wireless network, said method comprises the steps:
-
to initially classify a node (X, . . . , Y) as an anchor node (A) if a measure of uncertainty of the state estimate, such as a state estimate error covariance norm, is below a first threshold value, or else classify its node (X, . . . , Y) as a non-anchor node, giving rise to a partial order of said nodes (X, . . . , Y); every time a node (X) receives a message from another node (Y), said node (X) performs the following steps; to check that said node (Y) precedes said node (X) in said partial order, wherein a node (Y) preceding a node (X) meaning that the node (Y) being located in an ordered sequence comprising both node (X) and node (Y) and being located closer than node (X) to an initial anchor node in said sequence; to apply a recursive filter on the current state and measurement, if node (Y) precedes node (X), giving an updated state; and to update said status of said node (X) in said partial order; and
said method also comprises the step;to repeat the above steps, giving the position and orientation of each node (X, . . . , Y); and
in that for said method at least one estimation is chosen from the group consisting of the two estimations,position estimation, where the measurement equation system comprises vectors y and x related as Gy=Gx, where G is a matrix, where y is a vector comprising a position coordinate estimate for node (Y) received from said node (Y), and where x is a state vector comprising a position coordinate for node (X), and orientation estimation, where the measurement equation system comprises vectors z and x related as z=Hx, where H is a matrix, where z is a measurement vector comprising an angle of arrival for node (X), and where x is a state vector comprising an orientation coordinate and an angle of arrival for node (X). - View Dependent Claims (11, 12, 13, 14, 15, 16, 17, 18, 19)
-
Specification